jaeger changed the topic of #crux to: CRUX 3.7 | Homepage: https://crux.nu/ | Ports: https://crux.nu/portdb/ https://crux.ninja/portdb/ | Logs: https://libera.irclog.whitequark.org/crux/
tilman has quit [Ping timeout: 255 seconds]
tilman has joined #crux
zorz has left #crux [WeeChat 4.1.1]
disapper3nce has quit [Quit: leaving]
disapper3nce has joined #crux
disapper3nce has quit [Client Quit]
disapper3nce has joined #crux
ppetrov^ has joined #crux
deltahotel has joined #crux
deltahotel has quit [Quit: deltahotel]
groovy2shoes has quit [Ping timeout: 260 seconds]
remiliascarlet has quit [Ping timeout: 260 seconds]
groovy2shoes has joined #crux
remiliascarlet has joined #crux
remiliascarlet has quit [Read error: Connection reset by peer]
remiliascarlet has joined #crux
<Guest9824> hello
<SiFuh> Hello
<Guest9824> what would be the safest way to build packages for the machine itself that will only run a simple X and qemu.
<Guest9824> i have looked into fakeroot and pkgmk user making through prt-get but i don't think this is safe enough.
<Guest9824> anyone using separate qemu virtual machine to build packages for host machine?
<SiFuh> I myself used VM;s
<Guest9824> honestly, i love minimalism and that is why im here.
<SiFuh> Yes, I would install core ports only from the installation CD and every port built was from that.
<Guest9824> thank you, think i will do exactly that.
<Guest9824> have managed to build the initramfs, i was missing some kernel modules.
<Guest9824> CONFIG_BLK_DEV_RAM|CONFIG_BLK_DEV_INITRD|CONFIG_BLK_DEV_LOOP|CONFIG_MD|CONFIG_BLK_DEV_DM|CONFIG_DM_CRYPT|CONFIG_CRYPTO_USER|CONFIG_CRYPTO_USER_API_HASH|CONFIG_CRYPTO_USER_API_SKCIPHER|CONFIG_CRYPTO_AES|CONFIG_BLK_DEV_NVME|CONFIG_CRYPTO_XTS|CONFIG_CRYPTO_WP512|DAX
<Guest9824> as these are always required to boot, made them as built-ins and not modules
<Guest9824> thanks to SiFuh and ukky for helping me out go in the right direction.
<Guest9824> SiFuh: when you say 'I myself used VM;s' - why are you not using them anymore?
<SiFuh> I used a dedicated box for building. However, it has a problem now. It keeps shutting down.
<Guest9824> i see. thanks for the information
<Guest9824> SiFuh: one more question if it's not a problem.
<SiFuh> OpenBSD nfs server has no problem. 6:43PM up 206 days, 18:29, 2 users, load averages: 4.30, 4.17, 4.20
<Guest9824> my initramfs now boots, however, i'd like to use /dev/disk/by-uuid in initramfs rather than /dev/nvme0n1 and /dev/nvme1n1, for this to work it should be enough to create device with same major and minor block numbers or with 'cp -a' and add them to initramfs?
<Guest9824> ie, /dev/disk/by-uuid/bbcc1b73-1277-4515-9096-19fdbcfaae63 -> ../../dm-1
<Guest9824> i want to be able to switch nvme0 and nvme1 disks in their slots and always boot correct disk.
<SiFuh> Shouldn't that be a boot manager/kernel thing?
<Guest9824> dm0, 254 major 0 minor, dm1 254 major 1 minor
<SiFuh> I use
<SiFuh> root=/dev/mapper/machupo-root rw rd.auto=
<Guest9824> what do you mean, like kernel parameter thing?
<SiFuh> Your disks are encrypted right?
<Guest9824> yes
<Guest9824> i use this; /bin/cryptsetup luksOpen /dev/nvme0n1p2 crypt || exec /bin/sh
<SiFuh> Then use the name you chose root=/dev/mapper/machupo-root is what I chose.
<Guest9824> hmmm i don't think we're on the same line
<Guest9824> this is my append line for syslinux that boots
<SiFuh> I can swap out my disks and it still boots it
<Guest9824> APPEND luks enc_root=/dev/nvme0n1p2 root=/dev/mapper/crypt rw quiet
<SiFuh> Mine is APPEND root=/dev/mapper/machupo-root rw rd.auto=1 net.ifnames=0 quiet
<Guest9824> i'll need to read about rd.auto=1
<SiFuh> luks enc_root= <-- I do not use this
<Guest9824> is rd.auto something specific to dracut?
<Guest9824> because it appears it is, and i don't use dracut for my small initramfs
<SiFuh> It's a kernel option
<Guest9824> im unable to find anything related to rd.auto about kernel
lavaball has joined #crux
<SiFuh> rd.auto rd.auto=1 enable autoassembly of special devices like cryptoLUKS, dmraid, mdraid or lvm. Default is off as of dracut version >=024.
<SiFuh> Hmm maybe it is. Let me check
<Guest9824> got it. but why is dracut mentioned everythwere.
<SiFuh> Seems you are correct.
<Guest9824> hmmm...
<Guest9824> im not sure if using udev in initramfs should be the right way to deal with this problem, in case dm-0 is not always the first/faster/main slot and dm-1 being the second/slower/additional slot
<SiFuh> What about changing the root to a UUID?
<SiFuh> enc_root=UUID=bbcc1b73-1277-4515-9096-19fdbcfaae63
<SiFuh> Something like that? Is that a feature?
<Guest9824> let me look, seems possible
<SiFuh> Also seems a PARTUUID could exist too
<Guest9824> appears it is a feature. however how does it get the UUID if /dev/disk/by-uuid/ does not exist and the initramfs does not have udevd?
<SiFuh> Maybe it doesn't use /dev/disk/by-uuid/ ?
<Guest9824> maybe :) just trying to figure out. obviously need to do some reading.
<SiFuh> I saw the enc_root once before in this channel. I have never seen it anywhere else
<SiFuh> So I am kind of curious about it
<Guest9824> i always used enc_root. this is a text i wrote some time ago and how i build my initramfs; https://oshi.at/vVZW
<Guest9824> of course it can be shortened, it was written for other people to understand more
<Guest9824> i'll look into enc_root=UUID= further and try to figure this out.
<SiFuh> I only use dracut so not sure about it. I will read this link you posted. Looks okay
<Guest9824> thanks for the help, will come back with results about this enc_root=UUID thing
<SiFuh> Got to love gentoo guys
<Guest9824> :-)
<Guest9824> well, until somebody answers in #kernel, lets try it the simple stupid way by just trying enc_root=UUID :)
<SiFuh> Two entries in syslinux ;-)
<Guest9824> btw. i'm also interested in some more technical answer how kernel looks for these devices, it's not only if it works or not :)
tarxvfz has joined #crux
<SiFuh> Guest9824: Then you maybe want to read the answer
<Guest9824> +1 appreciate
<SiFuh> It's not exactly what we want but it says "It does not. The kernel uses device nodes."
lavaball has quit [Remote host closed the connection]
remiliascarlet has quit [Ping timeout: 255 seconds]
remiliascarlet has joined #crux
remiliascarlet has quit [Ping timeout: 240 seconds]
<cruxbot> [opt.git/3.7]: hpcups: update to 3.23.12
<cruxbot> [opt.git/3.7]: poppler: update to 23.12.0
remiliascarlet has joined #crux
zorz has joined #crux
tarxvfz has quit [Ping timeout: 240 seconds]
tarxvfz has joined #crux
joe9 has quit [Quit: leaving]
pitillo has quit [Ping timeout: 256 seconds]
pitillo has joined #crux
_whitelogger has joined #crux
<SiFuh> jaeger: whitequark finally active ;-)
pitillo has quit [Ping timeout: 245 seconds]
pitillo has joined #crux
tarxvfz has quit [Remote host closed the connection]
tarxvfz has joined #crux
zorz has quit [Quit: WeeChat 4.1.1]
zorz has joined #crux
zorz has quit [Client Quit]
<SiFuh> Anyway, the channel '#crux-social' is officially live where crux users can just gossip
zorz has joined #crux
<SiFuh> So if jue (No offense :-P) and others complain about offtopic then send them there
_whitelogger has joined #crux
lavaball has joined #crux
<cruxbot> [contrib.git/3.7]: wasi-compiler-rt: updated to 17.0.6
<cruxbot> [contrib.git/3.7]: wasi-libc++: updated to 17.0.6
<cruxbot> [contrib.git/3.7]: firefox: updated to 120.0.1
tarxvfz has quit [Remote host closed the connection]
remiliascarlet has quit [Ping timeout: 252 seconds]
remiliascarlet has joined #crux
joe9 has joined #crux
tarxvfz has joined #crux
zorz has quit [Quit: WeeChat 4.1.1]
zorz has joined #crux
<Guest9824> Unkj
<Guest9824> oops
<SiFuh> Heh
ivandi has quit [Quit: WeeChat 4.1.1]
ivandi has joined #crux
zorz has quit [Quit: WeeChat 4.1.1]
SiFuh has quit [Remote host closed the connection]
SiFuh has joined #crux
zorz has joined #crux
zorz has quit [Quit: WeeChat 4.1.1]
zorz has joined #crux
tarxvfz has quit [Quit: tarxvfz]
braewoods__ has quit [Read error: Connection reset by peer]
braewoods_ has joined #crux
groovy2shoes has quit [Remote host closed the connection]
groovy2shoes has joined #crux
<cruxbot> [opt.git/3.7]: c-ares: 1.22.1 -> 1.23.0
<cruxbot> [opt.git/3.7]: nodejs: 21.2.0 -> 21.3.0
<cruxbot> [contrib.git/3.7]: libmediainfo: 23.10 -> 23.11
<cruxbot> [contrib.git/3.7]: mediainfo: 23.10 -> 23.11
<cruxbot> [contrib.git/3.7]: notmuch: 0.38.1 -> 0.38.2
<cruxbot> [contrib.git/3.7]: p5-business-isbn-data: 20231125.001 -> 20231130.001
<cruxbot> [contrib.git/3.7]: papirus-icon-theme: 20231101 -> 20231201
<cruxbot> [contrib.git/3.7]: python3-frozendict: 2.3.9 -> 2.3.10
<cruxbot> [contrib.git/3.7]: python3-pyqt-builder: 1.15.3 -> 1.15.4
<cruxbot> [contrib.git/3.7]: python3-virtualenv: 20.24.7 -> 20.25.0
<cruxbot> [contrib.git/3.7]: sip: 6.7.12 -> 6.8.0
<cruxbot> [contrib.git/3.7]: ttf-noto: 23.11.1 -> 23.12.1
ukky has quit [Ping timeout: 264 seconds]
ukky has joined #crux
<farkuhar> ukky, here's my simple-minded attempt at implementing Mark Rosenstand's 15-year-old suggestion: https://dpaste.com/6FMD83MJD
<farkuhar> tested once, as a regular user with write-permission on the repo passed as argument. Will need further testing to see how it handles multiple repos and drivers.
<ukky> farkuhar: I will test your patch, observe results and report back when done.
<ppetrov^> farkuhar, is the idea that ports and prt-get eventually get "merged"?
<farkuhar> ppetrov^: that's one way to interpret https://crux.nu/Wiki/PkgutilsIdeas#ntoc8
<farkuhar> But as noted, it's a 15-year-old proposal that has been left dormant for a while.
<ppetrov^> yep
<ppetrov^> in a way, i kinda like the separation
<ppetrov^> but sth like prt-get sync would be nice
<ppetrov^> pkg-get sync exists
groovy2shoes has quit [Ping timeout: 268 seconds]
<farkuhar> ppetrov^: pkg-get sync was actually a strong motivation. The more parallelism between the commands, the less likely it is that muscle memory will produce a nonsensical command.
groovy2shoes has joined #crux
<ppetrov^> yep, makes sense
<farkuhar> speaking of pkg-get ... after the breakage stemming from the icu update, r0ni inquired about adding the -fr option (force reinstall) to pkg-get. I'm not sure how many CRUX users would take advantage of it, but there's no harm in letting pkg-get do something intelligent when it sees the -fr flag.
<ppetrov^> i would, for sure, benefit
ppetrov^ has quit [Quit: Leaving]
groovy2shoes has quit [Remote host closed the connection]
groovy2shoes has joined #crux